🔬 physics

Receiver-Limited Fisher Information for Quantum Sensors under Realistic Readout

This paper introduces a receiver-limited Fisher information framework that separates state-level sensitivity from channel and readout constraints, revealing how realistic experimental factors like measurement errors and calibration uncertainty fundamentally alter the performance and optimal operating conditions of quantum sensors.

Original authors: Juan Sumaya-Martinez, Omar Olmos-Lopez, Cristian Adán Mojica Casique, Jorge Mulia Rodriguez

Published 2026-10-26

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Juan Sumaya-Martinez, Omar Olmos-Lopez, Cristian Adán Mojica Casique, Jorge Mulia Rodriguez

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the world of precision measurement, scientists have long chased the ultimate limit of how accurately they can know something, from the position of a mirror to the strength of a magnetic field. For decades, the theoretical guide for this pursuit has been a concept called quantum Fisher information. Think of this as a measure of how much useful data is hidden inside a carefully prepared quantum state, a state that has been engineered to be incredibly sensitive to changes in the environment. The prevailing assumption was that if you could create a state with a high amount of this hidden information, your sensor would automatically be superior to any classical device. However, this view often ignores the final, critical step of the process: the actual machine used to read the data. Just as a high-resolution photograph is useless if the camera lens is blurry or the film is damaged, a perfect quantum state is of little value if the equipment reading it cannot extract the information it holds.

This gap between the potential of a quantum state and the reality of the equipment reading it is the focus of a new study by researchers Juan Sumaya-Martínez, Omar Olmos-López, Cristian Adán Mojica Casique, and Jorge Mulia-Rodríguez. They argue that the true performance of a quantum sensor is not determined solely by the quality of the probe, but by the limitations of the receiver. The team developed a framework to calculate the "accessible Fisher information," which represents the maximum amount of data that can actually be pulled out of a system by a specific, imperfect detector. Their work shifts the focus from asking "how good is the state?" to "how much of that goodness can our specific machine actually see?"

The researchers applied this new way of thinking to several common types of quantum sensors, revealing that the optimal settings for a device often change when you account for real-world errors. In one example involving a magnetic sensor that uses atoms to measure field strength, they found that the best time to run the measurement depends entirely on how you count your resources. If you are limited by the number of times you can run the experiment, the ideal duration is different than if you are limited by the total time you have available. This might seem like a minor detail, but it means that a sensor designed without considering the specific constraints of the lab could be running at the wrong speed, missing out on its best possible precision.

The study also looked at sensors that use entangled groups of particles, known as NOON states, which are theoretically capable of extreme precision. The researchers showed that these powerful states are fragile when it comes to the quality of the readout. If the detector has even a small amount of noise or reduced contrast, the advantage of using a large, complex entangled state can vanish. In fact, they found that there is a specific point where adding more particles to the entangled group actually hurts performance because the detector cannot keep up with the complexity. This means that simply building a bigger quantum state does not guarantee a better sensor; the state must be matched to the capabilities of the receiver.

Perhaps the most striking finding comes from sensors that use "squeezed" light, a technique that reduces noise in one property of light at the expense of increasing it in another. In an ideal world, you would want to squeeze the light as much as possible to get the most sensitivity. However, the researchers discovered that if your measurement equipment is slightly misaligned, squeezing the light too much becomes counterproductive. The extra noise from the wrong angle overwhelms the benefit of the squeezing. They derived a simple rule that tells engineers exactly how much squeezing is useful based on how well their equipment is aligned. If the alignment is not perfect, there is a specific, finite amount of squeezing that is best, and going beyond that point actually makes the sensor worse. This rule also applies to cases where the alignment is not perfectly known; the researchers showed how to choose a robust setting that works well even with uncertainty, turning a calibration tolerance into a direct design instruction.

The paper further explored how errors in reading the data, such as misidentifying the state of a particle, can shift the best temperature or operating point for a sensor. In a two-level system used for thermometry, a small error in the detector meant that the temperature at which the sensor worked best was no longer the same as the theoretical ideal. Similarly, when trying to measure two different properties at once, the researchers showed that if your detector can only see one combination of those properties at a time, you must split your measurement time between different settings to get the best overall result. This turns a theoretical limitation into a practical instruction on how to allocate time and resources.

Ultimately, this work provides a bridge between the abstract mathematics of quantum states and the concrete engineering of real devices. It demonstrates that the path to a better sensor is not just about creating more exotic quantum states, but about designing the measurement process to match the hardware that will be used. By explicitly accounting for the limitations of the receiver, the researchers offer a set of design rules that allow scientists to optimize their sensors for the real world, ensuring that the theoretical advantages of quantum mechanics are not lost in the translation to a working instrument. The study confirms that the future of quantum sensing lies not just in the preparation of the probe, but in the careful alignment of the entire chain from the quantum state to the final data point.

Technical Summary: Receiver-Limited Fisher Information for Quantum Sensors under Realistic Readout

Problem Statement
Quantum sensing is traditionally framed around the Quantum Fisher Information (QFI), which quantifies the ultimate precision limit encoded in a parameter-dependent quantum state. However, a critical gap exists between the information theoretically stored in a state and the information actually extractable by a physical receiver. While QFI assumes ideal measurements, real-world sensors are constrained by finite readout visibility, detector noise, calibration errors, and specific measurement architectures. This paper addresses the disconnect where a probe may possess a high QFI, yet the implemented sensor fails to realize a quantum advantage due to receiver limitations. The central problem is to determine how to optimize probe settings (e.g., interrogation time, squeezing strength) and operating points not just to maximize state-level sensitivity, but to maximize the Fisher information that survives the channel and is accessible to a specific, physically admissible receiver family.

Methodology
The authors propose a framework that explicitly separates the sensing process into three stages: generation (state preparation), preservation (channel noise/decoherence), and extraction (receiver measurement).

Receiver-Constrained Fisher Information ( $F_S$ ): Instead of optimizing over all possible Positive Operator-Valued Measures (POVMs), the authors define the accessible Fisher information as $F_S(\theta) = \sup_{M \in S} F_C(M)(\theta)$ , where $S$ is the family of physically admissible measurements for a given hardware.
Information Budgeting: The paper introduces a bookkeeping identity to track information loss:
- $\eta_{state} = F_Q^{post} / F_Q^{pre}$ : Survival of information through the channel.
- $\eta_{rec} = F_S / F_Q^{post}$ : Extractability by the receiver.
- $\eta_{op} = \eta_{state} \eta_{rec}$ : End-to-end information survival.
  This allows for the diagnosis of whether performance losses stem from state preparation/channel engineering or receiver architecture.
Resource Normalization: The analysis distinguishes between information per shot and information accumulated over a fixed total resource (time or total photon number), demonstrating that the "optimal" operating point depends on how the resource cost is defined.
Analytic Modeling: The framework is applied to specific models using closed-form derivations:
- Ramsey and NOON interferometry with visibility constraints.
- Squeezed-state displacement sensing with homodyne misalignment and isotropic loss.
- Two-level thermometry with assignment errors.
- Multiparameter qubit sensing with measurement rank deficiency and nuisance parameters.

Key Results
The paper derives several closed-form design rules and operational limits that differ from standard QFI-based predictions:

Ramsey Sensing: The optimal interrogation time ( $t^*$ ) is not unique. If each interrogation is an equal-cost resource, $t^* = 1/\Gamma$ (where $\Gamma$ is the dephasing rate). However, if the total interrogation time is fixed, $t^* = 1/(2\Gamma)$ . Finite readout visibility ( $V_R$ ) scales the classical Fisher information (CFI) by $V_R^2$ but does not shift the optimal time.
NOON Interferometry: For a receiver-limited NOON state with visibility $V_N$ , the operational quantum advantage over a resource-matched shot-noise benchmark requires $N V_N^2 > 1$ . If visibility decays phenomenologically as $V_N = e^{-\alpha N}$ , there exists an optimal photon number $N^* = \alpha^{-1}$ , beyond which increasing the probe size reduces the extractable information.
Squeezed-State Sensing:
- Fixed Misalignment: For a fixed homodyne-axis error $\phi$ , increasing squeezing ( $r$ ) indefinitely is counterproductive. There is a finite optimum squeezing strength given by $e^{2r^*} = |\cot \phi|$ .
- Bounded Uncertainty: If the alignment error is bounded by $|\phi| \le \Delta$ , a minimax design rule yields a robust squeezing setting of $r_{rob} = \frac{1}{2} \ln[\cot \Delta]$ .
- Isotropic Loss: In an isotropic pure-loss model, attenuation lowers the attainable Fisher information but does not shift the optimal squeezing point derived from the misalignment error.
Thermometry: Assignment errors in binary detectors shift the preferred operating point (temperature) rather than simply scaling the sensitivity. The optimal $x = \Delta/(k_B T)$ decreases as the error probability $\epsilon$ increases.
Multiparameter Sensing: A single-bit measurement on a two-parameter qubit is rank-deficient, resolving only one local combination of parameters. The paper shows that "measurement splitting" (allocating a fraction $f$ of shots to one basis and $1-f$ to another) can restore access to both directions. The worst-direction accessibility is maximized at a balanced allocation ( $f^* = 1/2$ ).

Significance and Claims
The authors position this work not as a replacement for QFI or a new quantum bound, but as a necessary "hardware-facing design layer" between state-level sensitivity and experimental reality.

Design Philosophy: The paper argues that state preparation, parameter encoding, and information extraction are distinct stages. Optimizing for QFI alone is insufficient when the receiver is constrained. The proposed framework provides "closed-form design rules" for realistic quantum metrology, translating calibration tolerances and receiver limitations directly into probe parameters (e.g., target squeezing levels).
Diagnostic Utility: The ratio $\eta_S = F_S / F_Q^{post}$ serves as a diagnostic tool. A low value indicates that the receiver architecture or calibration is the bottleneck, suggesting that engineering efforts should focus on the measurement stage rather than further improving the quantum state.
Operational Advantage: The paper emphasizes that a quantum advantage is only realized if the accessible Fisher information ( $F_S$ ) exceeds the best classical Fisher information ( $F_{cl}^{best}$ ) under the same resource constraints. This condition ( $F_S > F_{cl}^{best}$ ) is distinct from the condition $F_Q > F_{cl}^{best}$ .

In summary, the paper provides a unified, receiver-constrained formulation for quantum sensor design, demonstrating that the "optimal" operating point is often a function of the specific hardware limitations and resource definitions, rather than an intrinsic property of the quantum state alone.

Technical Summary: Receiver-Limited Fisher Information for Quantum Sensors under Realistic Readout

More like this